Journal of Physics: Conference Series

PAPER • OPEN ACCESS Related content

- Ethnomathematics: The use of multiple
Ethnomathematics elements in Batik Bali using linier regression Y = b1X1 + b2X2 + e in
traditional house construction Saka Roras
backpropagation method in Songan Village
J B Darmayasa, Wahyudin and T Mulyana

- Modification of chitosan by swelling and

To cite this article: Mei Lestari et al 2018 J. Phys.: Conf. Ser. 1022 012012 crosslinking using epichlorohydrin as
heavy metal Cr (VI) adsorbent in batik
industry wastes
B Hastuti, A Masykur and S Hadi

- The role of ethnomathematics in West

View the article online for updates and enhancements. Java (a preliminary analysis of case study
in Cipatujah)
Dianne Amor Kusuma, Stanley P.
Dewanto, Budi Nurani Ruchjana et al.

This content was downloaded from IP address 140.213.14.68 on 12/03/2019 at 23:35

ICoSMEE IOP Publishing
IOP Conf. Series: Journal of Physics: Conf. Series 1022 (2018)
1234567890 ‘’“” 012012 doi:10.1088/1742-6596/1022/1/012012

Ethnomathematics elements in Batik Bali using

backpropagation method

Mei Lestari1, Ari Irawan2, Wanti Rahayu3 and Ni Wayan Parwati4

1234
University Indraprasta PGRI, Jl. Raya tengah No.80, Jakarta Timur 13760,
Indonesia

Corresponding email: *1mei_lestari@unindra.ac.id, 2mascan_89@yahoo.com,

3
wanti_reiku@gmail.com, 4ni_wayanparwati@unindra.ac.id

Abstract. Batik is one of traditional arts that has been established by the UNESCO as
Indonesia’s cultural heritage. Batik has varieties and motifs, and each motifs has its
own uniqueness but seems similar, that makes it difficult to identify. This study aims
to develop an application that can identify typical batik Bali with etnomatematics
elements on it. Etnomatematics is a study that shows relation between culture and
mathematics concepts. Etnomatematics in Batik Bali is more to geometrical concept in
line of strong Balinese culture element. The identification process is use
backpropagation method. Steps of backpropagation methods are image processing
(including scalling and tresholding image process). Next step is insert the processed
image to an artificial neural network. This study resulted an accuracy of identification
of batik Bali that has Etnomatematics elements on it.

1. Introduction
Indonesia has varied cultural diversity; one of them is Batik which known as an icon of Indonesia.
Almost every region in Indonesia has Batik culture and they have unique characteristics that vary in
each region. As a form of International appreciation of Batik, UNESCO (United Nations Educational,
Scientific and Cultural Organization) officially crowned Indonesian Batik as World Culture Heritage
on October 2nd 2009 in France. World culture heritage is only specializing on Batik Tulis which
became an authentic characteristic of Indonesian Batik and not printed Batik. Is mentioned more than
181 batik motifs. The amount does not include a variety of local batik motif continues to grow [1].
Batik is a cultural heritage not because of the batik itself but because of the art in making the batik [2].
Marvin and Wahyuni et.al. said that culture is defined as a whole aspects of human life in society,
acquired by way of learning, include thoughts and behavior [3]. Wahyuni et.al., states that culture is
something that will affect the level of knowledge including the system of ideas or ideas contained in
human mind, so that in everyday life, the culture is abstract.
Meanwhile the embodiment of culture is the objects created by human as cultured creatures, such
as behaviour and objects that are real, for example; patterns of behaviour, language, life tools, social
organization, religion, art, and et cetera, which intended to help human in social life.

Content from this work may be used under the terms of the Creative Commons Attribution 3.0 licence. Any further distribution
of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.
Published under licence by IOP Publishing Ltd 1
ICoSMEE IOP Publishing
IOP Conf. Series: Journal of Physics: Conf. Series 1022 (2018)
1234567890 ‘’“” 012012 doi:10.1088/1742-6596/1022/1/012012

From the understanding above, it can be concluded, culture is as a manifestation of habits, objects
and tradition which is in heritage from generation to generation which still used by the inheritors.
Batik is a traditional fabric that became one of the nation's cultural richness of Indonesia[1]. Batik has
a long history, which began before the era of Raden Wijaya (1294 -1309), the first King of Majapahit.
During its development, style, motif, and colour of batik is also influenced by the culture from outside,
such as Hinduism, Islam, Netherlands, China, and Japan In the book "Spirits of Batik Indonesia" is
mentioned more than 181 batik motifs. Ethnomatematics research is suitable to do in Indonesia which
has a lot of culture. Despite the progress of the times happening, many ethnic in Indonesia who
survived by clinging to the doctrine of his ancestors[4].
Batik is pictorial fabric (patterned or fermented) which is made in certain way; first written and
scratched with candles, followed by colouring with tarum and soga. Bali is a province in Indonesia
[5]. Capital district in Bali is Denpasar. Bali is also the name of the island in this area. Wastika states
in Balinese cultural which most population is Hinduism, has their cultural character built since in the
fetus until death[6]. Their lives are fulfilling with religious and traditional ceremonies to be done.
From the understanding above it can be conclude that Balinese batik is tangible artworks or fabric that
have special motif which is influence by Hindu doctrine which exist in Bali.
Ethno-mathematics is a study about math that calculate cultural considerations where math arise by
understanding the system of the math that they use[3]. Ethno-mathematics is math which is grow and
evolve in a certain culture[7]. Etnomatematics is a concept, knowledge, study, or approaches that
associate mathematics with culture [8].The presence of math that have cultural nuances will give big
contribution for school mathematics, because school is social institution which is different with others,
it increase the possibilities of some cultural socialization[9]. Ethno-mathematics defined as cultural
anthropology of mathematics and mathematics education. From the understanding above, it can be
concluded that ethno-mathematics is one of math-elements which is exist inside of the surrounding
society. Consciously or not, that cultural have mathematical elements which applied on daily life. The
ethnomathematician mediator plays a major role in this approach. It is his or her reflections on the
things that has been said and done, and the perceptual shifts that he or she undergoes during the
investigation, that keep the dialogue going[10].
Based on those things, need a deeper identification in order to increase the devotion of batik culture
especially for Indonesia itself and worldwide. In this case for some region which hasn’t exposed yet
from the batik culture is Bali. Worldwide only knew that Bali is a place for tourist destination for
domestic or international tourist. Besides, there are thing stored well and maintained; Balinese Batik
which had vary variety and motif that is different with others regions especially Java. Balinese batik
have pattern that used ethno-mathematic, that unconsciously use mathematical elements.
Subjects that matter on ethno-mathematic are:
a. Emblems, concepts, principals, and mathematical skills which exist on national group, tribe, or
other communities.
b. The differences and similarity on mathematically things between each group and also factors
behind every differences and similarity.
c. Interesting or specific thing between one group and several groups, for example: the way to
think, behave, speak, and et cetera that related with math.
d. Every aspect on social life is related with mathematic, for example: financial literacy and
economic awareness, social justice, cultural awareness, and political awareness.
There are many people who doesn’t know mathematical elements on batik’s motif therefore it
needs deeper study about those elements. In addition to overcome the ignorance of the Balinese people
in particular about the motive and variety of batik in Bali it is important to have a method that can be
used to identify the motive of Balinese Batik. This is what underlies the use of back propagation
algorithms in recognizing Balinese motives.
Back-propagation algorithm itself is one of the most common methods from artificial neural
networks for the pattern identification and had high accuracy. Back-propagation trainee the networks
to balance between network capability to identify the pattern used and also the capability to give the

2
ICoSMEE IOP Publishing
IOP Conf. Series: Journal of Physics: Conf. Series 1022 (2018)
1234567890 ‘’“” 012012 doi:10.1088/1742-6596/1022/1/012012

accurate response for similar pattern (but different) with pattern that use on training period[11]. On
training period of back-propagation algorithm there are three phases: forward propagation, backward
propagation and weight changes. First, propagate the input signal to the hidden screen with activation
functions which determined. Second, back-propagation phase, every output unit which already done
on the feed forward phase will receive pattern that targeted that related with input pattern to calculate
the mistakes numbers. Third, weight modifications in order to decrease the numbers of mistake.

2. Methods
This research is use qualitative methods with explorative approach to reveal ethno-mathematical
elements in Batik’s motif. Research development of the application maker of Batik motif identifies
and contains ethno-mathematic elements. Sample which used in this research is five motifs that
contain ethno-mathematics element inside.

3. Result and Discussion

Balinese batik had unique characteristics; vary of the motifs, colours and also fabrics. The name of the
Balinese batik also gained because of the varies of the motifs, for instance; Balinese batik buketan,
merak abyorhokokai, singabarong, ulamsari mas, jagatan pisang, et cetera. Each motif drawn in order
to accentuate a characteristic that reminds us to Balinese culture which is really sacred, mythology,
tradition, culture, and the artworks. In Bali the use of Balinese Batik still preserve by still use Batik in
every traditional and religious ceremonies.

Figure 1. Motif Abyorhokokai

On abyorhokokii motif seen only on the rectangular frame that surrounds the image of the peacock.
There is only a reflective element, however it’s created with different colour between each flowers.
The most predominate on this motif is there are two peacock which is face-to-face. The meaning is
where the destination of human soul afterlife. From this illustration can be concluded that death is only
body damage, meanwhile the soul still alive run into the creator (God).

Figure 2. Motif Buketan

Buketan mean a bunch of flowers, and this name adopted by French language. The motif of this
Balinese batik is flowers pattern or small plants arranged along the fabric with an additional
decoration such as butterfly, hong, stork, and also vines that increase the values of the art from this
batik. Ethno-mathematic elements from this batik can be seen from the reflection of the geometry
element; black fabric and gold coloured flowers.

3
ICoSMEE IOP Publishing
IOP Conf. Series: Journal of Physics: Conf. Series 1022 (2018)
1234567890 ‘’“” 012012 doi:10.1088/1742-6596/1022/1/012012

Figure 3. Motif Jagatan Pisang

Some people called this motif as Balinese banana, this motif describes Balinese banana which has
been distilled. There are called as Balinese banana which mean “back again/retro”. This batik usually
given from someone to their loves one who will travel far away in order to make the loves one will
return. Based from the picture above, there are mathematical elements; curved lines with variety arch
and every arch are connected. Besides, there are many reflection elements and relatable between each
motifs

Figure 4. Motif Singa Barong

Barong lion motifs has a very few ethno-mathematical elements, because this batik is the two-
pieces image of Barong which different and also facing each other. It is only can be seen by geometry
element; circles and curved lines that facing each other. The meaning and philosophy of this batik is
one of mythological or magical animals, because in Balinese or Javanese culture, “barong” means
“magical”. The miracle of it is being can be seen by the existence of these elements: integration
between lion and tiger (body, foot, and eye), elephants (trunk), garuda (wings), dragon (grinning
mouth with tongue sticking out). The embodied animals in Barong lion not just because of it is
physical strengthen, but also because of the spiritual symbols.

Figure 5. Motif UlamSari Mas

Batik ulam sari mas motif is a very simple batik motif if viewed from the ethno-mathematical
element which exists on the batik. In this batik the patterns mostly fulfil with reflection of fish and
shrimp contradict each other when played. Ulamsari mas are symbol from the society who is
livelihood as a fisherman; fish also represented prosperity from people who live in coastal areas.
Data of the batik motif that has been analysed by ethno-mathematical elements will be process with
image processing; scaling, grayscale and tresholding stages. Those stages are done to obtain input with
binary image followed with learning artificial neural network back-propagation.

4
ICoSMEE IOP Publishing
IOP Conf. Series: Journal of Physics: Conf. Series 1022 (2018)
1234567890 ‘’“” 012012 doi:10.1088/1742-6596/1022/1/012012

3.1. Scaling
Image of Abyorhokokai batik motif measure 350 x 225 pixel, minimize into 50 x 50 pixel

Figure 6. Minimize Batik Abyorhokokai Motif Image

After the pixel size changes, then find the value of RGB every pixel. On the following table will
showed RGB value of Abyorhokokai batik motif only to the size 5 x 5 pixel.

Table1. Value of RGB Abyorhokokai Motif

Pixel 1 2 3 4 5
1 252,255,255 252,255,255 250,252,249 254,255,251 254,254,255
2 245,253,248 255,255,246 255,255,241 252,251,233 255,255,243
3 255,255,230 145,136, 105 156,147,108 165,156,113 148,136,96
4 255,255,233 157,145,107 157,145,97 155,142,89 253,137,85
5 255,255,228 147,138,105 85,74,29 116,104,54 170,154,105

3.2. Grayscale
After scaling, image will perform grayscale stage, in this stage image will turn into gray.

Figure 7. Citra Grayscale

Thereafter find the grayscale value of each pixel with formulation: Grayscale = (R + G + B) / 3
With the result that obtains grayscale value up to size 5 x 5 pixel as follows:

Table 2. Value of Grayscale Buketan Motif

Pixel 1 2 3 4 5
1 254 254 250,3 253,3 254,3
2 248,7 252 250,3 245,3 251
3 246,7 128,7 137 144,7 126,7
4 247,7 136,3 133 128,7 158,3
5 246 130 62,7 91,3 143

5
ICoSMEE IOP Publishing
IOP Conf. Series: Journal of Physics: Conf. Series 1022 (2018)
1234567890 ‘’“” 012012 doi:10.1088/1742-6596/1022/1/012012

3.3. Tresholding
Tresholding is process that convert image into black and white image or binary image.

Figure 8. Thresholding
Calculation into binary form with this following rule:
a. value of threshold range between 0-255 limit T = 128.
b. if the value of the pixel more or same with 128, then changes into 1, meanwhile if the value of the
pixel is less than 128, then changes into 0.

Table 3. Binary Value of Abyorhokokai Motif

Pixel 1 2 3 4 5
1 1 1 1 1 1
2 1 1 1 1 1
3 1 1 1 1 1
4 1 1 1 1 1
5 1 1 0 0 1

The output as shown in table 3 will be used as input in the learning of algorithm back-propagation.

3.4. Training
Input on training process is the result of treshloding process of batik abyorhokokai. Then, the motif
transformed into 1 dimension array.
Stages of learning process are:
a. Set the value of epoch, rate of understanding, minimum error. For example: in this calculation limit
that used are: epoch = 10 rate of understanding = 0.2, range 0- 1, in this case used value 0.2 as
learning ratio with minimum error = 0
b. Initialize the initial weights with random values with intervals -0.5 – 0.5
c. Loop is performed as long as loop value is less than the maximum of the iteration and the square of
error is greater than the minimum error value. Each data is looped counting step 4 and 5.
d. Feed-forward
First loop
First data
e. Back-propagation
On the second data do the same operation using the final result from the first data. This process
repeated until the maximum iteration.

3.5. Identification
Identification is process to recognize the testing image identification. For example, there are image of
Abyorhokokai batik motif scale 50 x 50 pixels which have been pre-processed and transformed from
matrix into one array.

6
ICoSMEE IOP Publishing
IOP Conf. Series: Journal of Physics: Conf. Series 1022 (2018)
1234567890 ‘’“” 012012 doi:10.1088/1742-6596/1022/1/012012

Identificationsteps are:
a. Do the learning process with feed-propagation and back-propagation. The learning process has
been done and obtained the weight of learning outcomes and the value of Y.
b. After the obtained the weight of the learning outcomes, do the identification process with feed-
forward-propagation, it will get the value of Y from the output of identification process.
c. After obtained the value of Y from the identification, the value of Y from identification compared
with value of Y which on the learning target, then the data will be obtained. Y from identification is
Y from calculation using the most optimal weight of W and V that obtained from the learning
process that is done repeatedly until the epoch is met, therefore Y from identification will approach
Y from learning process, so that the value of this Y is the matching data of learning and training.

3.6. Testing
Testing are aims to know the accuracy of the image recognition of Balinese batik motif from the back-
propagation methods which applied on the simulation to identify the batik motif.
The performance testing plan that will run on this simulator is a test using various combinations of
learning parameters; it is shown in table 4.
Table 7. Testing Result of Parameter 2
Table 4. Testing Parameter
Motives name Target Result Label
Parameter Value Abyorhokokai 1,0000 1,0028 Identified
Epoch 10 Buketan 2,0000 2,0031 Identified
50 Jagatan 3,0000 3,0011 Identified
Understanding rate 0,1 Pisang
0,2 Singa Barong 4,0000 4,0003 Identified
Error Minimum 0,01 Ulam Sari 5,0000 4,3501 Un
Mas Identified
Akurasi=4/5*100%=80%
Table 5. Parameter combination
No Epoch Learning rate Minimum Error Table 8. Testing Result of Parameter 3
1 10 0,1 0,01 Motives name Target Result Label
2 10 0,2 0,01 Abyorhokokai 1,0000 1,3596 Identified
3 50 0,1 0,01 Buketan 2,0000 2,2063 Identified
4 50 0,2 0,01 Jagatan 3,0000 2,3949 Un
Pisang Identified
Table 6. Testing Result of Parameter 1 Singa Barong 4,0000 4,0024 Identified
Ulam Sari 5,0000 4,9951 Identified
Motives name Target Result Label Mas
Abyorhokokai 1,0000 1,0028 Identified Akurasi=4/5*100%=80%
Buketan 2,0000 2,0331 Identified
Jagatan 3,0000 3,0011 Identified Tabel 9. Testing Result Parameter 4
Pisang
Singa Barong 4,0000 4,0001 Identified Motives name Target Result Label
Ulam Sari 5,0000 4,3501 Un Abyorhokokai 1,0000 2,6637 Un Identified
Mas Identified Buketan 2,0000 2,0144 Identified
Akurasi=4/5*100%=80% Jagatan 3,0000 3,0773 Identified
Pisang
Singa Barong 4,0000 3,9109 Identified
Ulam Sari 5,0000 5,0780 Identified
Mas
Akurasi=3/5*100%=60%

7
ICoSMEE IOP Publishing
IOP Conf. Series: Journal of Physics: Conf. Series 1022 (2018)
1234567890 ‘’“” 012012 doi:10.1088/1742-6596/1022/1/012012

Accuracy Rate
1st 2nd 3rd 4th
Testing Testing Testing Testing
Accuracy 80% 80% 80% 80%
Accuracy rates = (80%+80%+80%+80%)/4= 80%

Application Design

Figure 9. Application Design

4. Conclusion and Suggestion

4.1 Conclusion
Based on the result that obtained from the research and adjusted with the purpose, then obtained the
conclusion, average accuracy rate is 80%, which mean the accuracy rate to identify the batik motif that
contained ethno-mathematical elements are really good.
4.2 Suggestion
a. in order to increase the performance of this system, the sampling of the batik images need to be
multipLied for the trainee of neural network system, to increase the accuracy of this system.
b. For the future work it is sugested to built an application to Identify bati’s motif that has
ethnomathematic element.

5. References
[1] R E Putra N, Suciati and A Y Wijaya 2011 Implementing Content Based Image Retrieval For
Batik Using Rotated Wavelet Transform And Canberra Distance in Bali International
Seminar on Science and Technology pp 1–5
[2] A H Rangkuti, Z E Rasjid and D J Santoso 2015 Batik image classification using treeval and
treefit as decision tree function in optimizing content based batik image retrieval
Procedia Comput. Sci 59 no Iccsci pp 577–83
[3] A Wahyuni, A A W Tias and B Sani 2013 Peran etnomatematika dalam membangun karakter
bangsa in Prosiding Jurusan Pendidikan Matematika FMIPA UNY. Yoyakarya:
Universitas Negeri Yogyakarta pp 113–18
[4] T Septianawati, Turmudi and E Puspita 2016 International Conference on Recent Trends in
Physics 2016 (ICRTP2016) in Journal of Physics: Conference Series 755 pp 1–7
[5] R U Amalia 2010 Motif batik Pekalongan  : studi dokumen koleksi museum batik Pekalongan
Seni dan Budaya 6(2) pp 125–41
[6] D N Wastika 2005 Penerapan konsep tri hita karana dalam perencanaan perumahan Di Bali J.
Pemukim. Natah 3(2) pp 72–7
[7] K R Puspadewi and I G N N Putra 2014 Etnomatematika di balik kerajinan anyaman Bali J.
Mat. 4(2) pp 80–9
[8] D A Kusuma, S P Dewanto and B Nurani 2016 The role of ethnomathematics in West Java (a
preliminary analysis of case study in Cipatujah) in Journal of Physics: Conference Series
pp 1–8

8
ICoSMEE IOP Publishing
IOP Conf. Series: Journal of Physics: Conf. Series 1022 (2018)
1234567890 ‘’“” 012012 doi:10.1088/1742-6596/1022/1/012012

[9] F S Sirate 2012 Implementasi etnomatematika dalam pembelajaran matematika pada jenjang
pendidikan sekolah dasar Lentera Pendidik.15(1) pp 41–54
[10] N A Adam 2010 Mutual interrogation: A methodological process in ethnomathematical research
Procedia - Soc. Behav. Sci. 8 pp 700–7
[11] J J Siang 2009 Jaringan Syaraf Tiruan & Pemrogramannya Menggunakan Matlab Yogyakarta:
Andi